You are on page 1of 6

Gear Measuring Machine

by NDG Method
for Gears Ranging from
Miniature to Super-Large
Masatoshi Yuzaki

This is an interesting concept and should be of While the conventional instruments go back to basics and
interest to your readers. measure normal to the tooth surface at the base circle tan-
Robert E. Smith gent, (the authors) NDG method has the probe moving in a
direction that is not normal to the surface. He therefore has
Robert E. Smith is president of R.E. Smith & Co. Inc., a to make a correction to all measurements involving the co-
gear consultancy in Rochester, NY. A mechanical engineer, sine of the transverse pressure angle. However, that is not
he has more than 60 years experience in the gear industry. a problem. Gear measurements on a CMM-type instrument
He is chairman of the AGMA Calibration Committee and was have to be corrected in a similar way to the surface normals.
AGMAs ISO delegate for that panel as well as for the AGMA When we established our National Gear Metrology labora-
Inspection and Handbook Committee. Since 1991, he has vol- tory at Oak Ridge, we had them measure artifacts by the first
unteered his services as aGear Technologytechnical editor. principles (TDG) method in order to compare to CMM-type
As Bob was the technical reviewer of this article, we believe measurements that required algorithms for probe corrections.
his comments regarding this papers relevance will be of in- We were satisfied that the results were good and the differ-
terest to readers. (The Editors) ences were insignificant. All artifacts today are measured by
the CMM instruments. In this case, the author did compare
This is an interesting concept and should be of interest measurements by both TDG and NDG systems, on the same
to your readers. As the author points out, it has several advan- gear, and showed the results to be the same. It was done on a
tages over the conventional TDG method. When reducing the gear of approximately 13-inch diameter, but I would expect
X axis movement while checking large-diameter gears, the good results on much larger gears also.In fact, the error, or
potential errors of probe positioning are reduced. Also, the uncertainty of measurement, should be (even) less on larger
instrument can be smaller in that direction. There are definite gears.
advantages to checking small-diameter internal gears, also.

Management Summary
A study was conducted on the development of a CNC gear measuring machine for measuring involute tooth profile
by a new measurement method. Involute tooth profile measurement has been done, until now, by almost always using
two-axis control in which the probe moves only in the X-axis direction synchronously with the gear rotation angle ().
In contrast, the newly developed measurement method uses three-axis control in which the probe moves along the line
of action under control in two orthogonal, axial directions (along the X and Y axes) synchronously with the gear rota-
tion angle ().
This new method enables high-accuracy measurement because the small X-direction movement of the probe reduces
the guaranteed accuracy range and minimizes movement of the probing head gravity center. As probe movement in the X-
direction is unaffected by gear outside diameter, the advantage of the new method over earlier ones is particularly relevant
to the measurement of super-large gears. While conventional measurement methods must use multiple probes to avoid
probe-tooth interference in the measurement of inner gears, the new method uses fewer probes in inner gear measurement
and eliminates the need for an automatic tool changer (ATC). In the case of a small-diameter inner gear (outside circle
diameter of 10mm or less), measurement of tooth profile, helix and pitch deviation can be completed with a one-time set-
ting. A CNC gear measuring system is developed using this new measurement method that provides numerous advantages
over conventional measuring systems.
continued
www.geartechnology.com March/April 2011 GEARTECHNOLOGY 55
Introduction
Almost without exception, conventional dedicated gear
measuring machines measure involute tooth profile by the
two-axis control method: the probe moves only along the X
axis, orthogonal to the axis of rotation synchronously with the
gear rotation angle () (Figs. 12). As it is based on the prin-
ciple of involute tooth profile generation, this conventional
method is quite simple and easy to understand.
However, the working positions in the hobbing machine,
gear shaper, gear grinder and other tooling are almost all near
the center of the gear, and these working positions are very
different from the conventional tooth profile measurement
positions. In the hobbing machine, the tooth cutting position
of the hob cutter moves along the gear line of action while
rotating synchronously with the gear blank (Fig. 3).
The author long questioned the reason for the significant
difference between the working positions during gear machin-
ing and the measurement positions during gear measurement
(tangent to the base circle). This current study grew out of an
intuition that it should work to control the probe movement at
the gear cutting position.
The author will use the term tangential direction gener-
ate method (TDG Method) for the ordinary measurement in
Figure 1Coordinate system of gear measuring machine.
which the probe moves only in the direction tangential to the
base circle (X-axis direction orthogonal to the axis of rota-
Y tion), and the term normal direction generate method (NDG
Method [patent applied]), for the new measurement method
in which the probe moves along the lines of action (X and Y
axes orthogonal to the axis of rotation).
Normal Direction Generate Method (NDG Method)
X
Amount of probe movement. As seen in Figure 3, the
NDG Method measures tooth profile as the probe is moved
Base Circle
along the line of action in the same way as the hob cutter dur-
ing tooth cutting. The probe moves in the direction of touch-
Figure 2Typical tooth form measurement. ing the base circle, and the principle of involute tooth profile
Figure 2Typical tooth form measurement
generation is exactly the same as in the TDG Method.
Measurement of a standard gear without any profile shift
(whose height of the involute tooth profile portion is one mod-
ule on both the addendum side and the dedendum side) by the
La Ld NDG Method gives La = Ld = m/tant (mm), where La and Ld
are the amounts of probe X-axis direction movement from the
Y axis passing through the gear center on the addendum side
Line of Action and the dedendum side, respectively; m is the module (mm),
Hob cutter
and t is the transverse pressure angle. When t = 20, La =
Ld 2.7 m. Measurement is the same for the right and left
tooth faces. Thus the amount of probe movement in the NDG
Method is independent of the number of gear teeth and diam-
eter of the reference circle.
X On the contrary, the amount of X-axis direction probe
movement in the TDG Method is approximately proportional
to the reference circle diameter (Fig. 4) and given by:
Base Circle
Ll = Lr = 0.5 z m (1 + 2 / z )2 cos 2 t
t A graph representing (Ll + Lr ) when m = 10 mm is shown
in Figure 5. The amount of X-axis direction probe movement
increases with the diameter of the reference circle. A machine
capable of measuring a super-large gear like that in Figure 6
using the TDG method is therefore very difficult to build.
Figure Figure
3Tooth form measurement
3Tooth by NDG
form measurement Method.
by NDG-Method Measurement error. While probe movement is solely in
56 GEARTECHNOLOGY March/April 2011 www.geartechnology.com
the X-axis direction during tooth profile measurement by the
TDG Method, it occurs in both the X-axis and Y-axis direc- Ll Lr
tions of the measuring machine in the NDG Method. The
profile measurement error caused by probe movement er-
ror during measurement therefore differs between the TDG
Method and the NDG Method. The nature of this difference X
was investigated.
Base Circle
In Figure 7, the probe being positioned at A indicates
tooth profile measurement by the TDG Method, and the probe
being positioned at B indicates tooth profile measurement by Figure 4Amount of probe movement by TDG Method.
the NDG Method. Figure 4Amount of probe movement by TDG-Method

Let us assume that maximum tooth profile measurement 3000

Amount of probe movement Ll +L r (mm)


error in the TDG Method, designated F-TDG, can be expressed
by: 2500

F TDG eX eL rb (1) 2000


where:
1500
eX is the maximum X-axis direction error of the probe aris-
1000
eX independently of its X axis position,
ing
e F NDG e X cos t eY sin t rb
eXL 500
e is the maximum X-axis direction error of the probe aris-
L due to large movement of the probe,
ing 0
e 100 1000 10000
L
Figure 5Amount of Reference
probe movement in(mm)
pitch diameter X-axis direction by
F NDG F TDG e X (cos angle
rotation eY sin t eL
t 1) error,
is the maximum gear TDG Method.
and:

rb is the radius of the gear base circle.


F TDG eX eL rb
F NDG F TDG e X (sin t cos t 1) eL
Let us further assume that maximum tooth profile mea-
surement error in the NDG Method, designated F-NDG, can be
expressed by:
F FNDG e X Fcos
NDG
t e0Y.28
TDG
sineX t eL rb (2)
where: F
F TDG
TDG eeXX eeLL
rrbb
eY is the emaximum
Y F Y-axis
TDG edirection
X eL error rof b the probe.
As the amount of probe movement
F NDG F TDG e X (cos t 1) eY sinis small inthe NDG
eL
t
eL
Method, eL is assumed to be negligibly small.
The maximumF NDG
F ee XX cos
NDG tooth
cos tt
profile eeYY sin
sin tt
measurement rrbb
errors by the
NDG Method and TDG Method are compared by taking the
F NDG e X cos t eY sin t rb
difference between Equation 1 and Equation 2 to obtain Equa- Figure 6TDG Method measuring machine for a super-big
tion 3: F NDG F TDG e X (sin t cos t 1) eL gear.
F NDG
F F TDG
NDG F
e X (cos
TDG e X
(cos tt 1 eeYY sin
1)) eeLL (3)
sin tt
L
F NDG
From Fthat
the fact TDGeY is (cos the
e Xabout 1) esize
t same Y sinas te ,eL
X
the following expressionF NDGholds:
F TDG 0.28 eX eL eX
eL eL eX eY sin t
F NDG
F NDG
F
F
TDG
TDG
e
e X
X
(sin
(sin tt

cos
cos tt

1
1 )
)
eeLL (4)

F NDG F TDG e X (sin t cos t 1) eL eY
when:
t = 20, B
F NDG
F F TDG
NDG F
0.28 eX
TDG 0.28 e X
eeLL X
A
F NDG F TDG 0.28 eX eL (5) e X cos t

eY This means that the tooth profile measurement error of the Base Circle
TDG Method is greater than that of the NDG Method when t
eL exceeds 0.28 eX .
To give a specific example: at eX of 0.2 mm, the TDG
eL
Method becomes greater in measurement error than the NDG Figure 7Measurement error by TDG Method (A) and NDG
Figure 7Measurement Error by TDG-Method (A) and NDG-Method (B)
eL continued Method (B).



www.geartechnology.com March/April 2011 GEARTECHNOLOGY 57
Table 1Specifications of the Developed NDG Measuring Machine
Test Mode profile, lead, pitch
for spur/helical gear, internal gear
with auto alignment system
Machine Size 4,300 x 1,200 x 3,400 mm
Weight 6,000 kg
Measurement 0.1 m
accuracy
Module 1.0 to 32 mm
Outer diameter 2,000 mm
(max)
Face width 1,500 mm
Gear (max)
Helix angle 65 deg
(max)
Shaft length 150 to 2,000 mm
Weight (max) 10,000 kg

Y indicates measurement of inner gear tooth profile by the TDG


Method. When a probe of small tip diameter is used, interfer-
B
ence occurs between the tooth tip and the probe stem. If an at-
tempt is made to avoid interference between the tooth tip and
A probe stem by enlarging the probe tip diameter, interference
X
will then occur between the probe tip and the opposite tooth
face. Although this problem can be overcome by removing
Tooth tip one side of the probe tip, it would require the probe orien-
Base Circle
tation to be reversed laterally when measuring the opposite
tooth face.
Figure 8Measurement of inner gear by TDG-Method (A) On the other handas shown in Figure 8Bmeasure-
and NDG-Method
Figure 8Measurement(B).
of inner gear by TDG-Method (A) and NDG-Method (B) ment by the NDG Method does not experience interference
between the tooth tip and probe, no matter how much probe
tip diameter is reduced. The left and right tooth faces can
therefore be measured with a single, small probe. The effi-
Line of Action
cacy of the NDG Method is therefore particularly evident in
the measurement of small-diameter, inner gears ( 10mm or
Reference Circle less).
B
Probe tip position. It should also be noted regarding mea-
A surement by the NDG Method that it differs from that of the
X
TDG Method not only in the direction of probe movement,
but alsodepending on the probe typein the initial probe
position.
Base Circle
t Figure 9 shows an example in which the probe has a
spherical tip. In the TDG Methodindicated by A in the
drawingthe center of the tip sphere is positioned on the X
eY axis, where the measurement is performed. In the NDG Meth-
Figure 9Measurement by sphere tip
Figure 9Measurement probe.
by sphere tip probe
odindicated by B in the drawingthe center of the probe
Method when eL becomes 0.056mm or larger. In a machine tip sphere is positioned at the intercept of the reference circle
for measuring super-large gears, it is extremely difficult to and the Y axis, and measurement is performed along the line
achieve a probe movement error of less than 0.056mm. The of action at pressure angle t.
NDG Method is therefore better than the TDG Method for Figure 10 shows an example in which the probe has a
measuring super-large gears. chisel type tip. In the TDG Methodindicated by A in the
Measurement of inner gear. When an inner gear is mea- drawingthe tip of the chisel is positioned on the X axis,
sured by the TGD Method, it becomes impossible to measure where the measurement is performed. In the NDG Method
tooth profile, helix and pitch deviation with a one-time setting indicated by B in the drawingthe axis of the probe stem is
because interference arises between the tooth face and probe, aligned with the Y axis and the chisel tip must be moved from
thereby causing frequent interruptions. the intercept of the reference circle and the Y axis toward the
In Figure 8, the case where the probe is positioned at A gear center by (dp / 2) tan t , where dp is the tip diameter of the
58 GEARTECHNOLOGY March/April 2011 www.geartechnology.com
chisel type probe.. Table 2Gear Dimensions for Test Measurement
NDG Measuring Machine
A measuring machine utilizing the newly developed Module 5 mm
NDG Method is shown in Figure 11. The specifications of Number of teeth 60
the developed measuring machine are provided in Table 1. Outer Diameter 329.5 mm
The X axis direction width of the measuring machine can be
Pressure Angle 20 deg
slimmed down considerably, compared with one adopting the
conventional system. Helix Angle 20 deg
In the measuring machine using the NDG Method, the Gear Width 50 mm
probe is controlled in orthogonal, two-axis (X and Y) direc-
tion at a given angle to move along the line of action. It should Table 3Profile Deviations by
therefore be noted that the tooth profile error output of the TDG-Method and NDG-Method
probewhich has sensitivity in the X axis directionis the Unit:mm
cosine (cos) of the transverse pressure angle. In other words,
Left Flank Right Flank
the tooth profile error must be multiplied by the displacement
Fa ffa fHa Fa ffa fHa
output (1/cosat). And, as mentioned previously, the probe TDG 3.8 1.7 -3.4 2.4 0.7 -2.4
must be initially positioned so that the measurement point
NDG 3.8 1.7 -3.4 2.4 0.7 -2.4
falls on the line of action.
Measurement Result
The dimensions of a gear for test measurement are shown
in Table 2.
NDG Method measurement is performed near the gear dp
center, analogous with the working positions in a gear man- tan t
2
ufacturing machine. This makes measurement possible in a Reference Circle
much shorter time than by the TDG Method. A comparison B
of measurement times using the same developed measuring A
X
machine showed that the NDG Method achieved a time re-
duction of 35% for profile measurement, compared with the
TDG Method. dp
Base Circle
Figure 12 and Table 3 show the results when tooth pro-
file measurement is conducted by the NDG Method and TDG
Method in the same developed measuring machine. The re-
sults show that there is no substantial difference in measure-
continued
Figure 10Measurement by chisel tip probe.
Figure 10Measurement by chisel tip probe

Figure 11Developed NDG measuring machine.

www.geartechnology.com March/April 2011 GEARTECHNOLOGY 59


ment accuracy between the NDG Method and TDG Method. High-accuracy tooth profile management is hard to
Conclusion achieve in the measurement of dies and molds used for
The following are some typical problems that arise when plastics, sintered metals, forgings and the like.
tooth profile is measured by the conventional TDG Method, In contrast, advantages such as the following are obtained
in which the probe moves only in the X axis direction: when tooth profile measurement is performed by the NDG
High-accuracy measurement cannot be anticipated Method, in which the probe moves along the line of action
in a measuring machine for large gears because under control in two orthogonal, axial directions (along the
the large movement of the probing head expands the X and Y axes):
guaranteed accuracy range and increases movement High-accuracy measurement is possible, even in a
of the probing head gravity center. measuring machine for large gears, because the
The X axis direction movement of the probe is small X axis direction movement of the probe
proportional to the gear reference circle diameter, reduces the guaranteed accuracy range and minimizes
making large-gear measurement time consuming. movement of the probing head gravity center.
When an inner gear is measured, it is often impossible As measurement is conducted near the gear center,
to measure tooth profile, helix and pitch deviation even a large gear with a large reference circle diameter
with a one-time setting because interference arises can be measured in a relatively short time.
between the tooth face and probe, thus necessitating In inner gear measurement, the fact that no interfer-
frequent interruptions. ence arises between the gear and the probe reduces
Numerous probes matched to different gear sizes are the number of probes required, eliminates the need
necessary. An automatic tool changer is therefore for an automatic tool changer, and minimizes probe
more often required, and probe calibration work calibration work.
increases. Measurement of tooth profile, helix and pitch devia-
Measurement of a small-module inner gear (minimum tion can be completed with a one-time setting,
outside circle diameter of 10mm or less) is difficult. even in the case of a small-diameter inner gear
(outside circle diameter of 10mm or less).
NDG Meas. Left A program enabling NDG Method measurement can be
incorporated into an existing CNC gear measuring machine.
When an NDG Method measuring machine was actually
TDG Meas. Left built, it was found that the X-axis direction width of the ma-
chine could be made slimmer than a conventional one. In ad-
dition, the tooth profile measurement results were found to be
substantially no different in accuracy from those by the con-
ventional system. The NDG Method was incorporated into a
conventional measuring machine and was confirmed to be a
tooth profile measuring method applicable to gears ranging
widely in sizefrom super-large to miniature.
TDG Meas. Right (In closing, the author wishes to express heartfelt grati-
tude to the team members who built the measuring machine.)

NDG Meas. Right


Figure 12Profile measurement results by TDG Method and
NDG Method.

Masatoshi Yuzaki is president of To-


kyo Technical Instruments, a company
he founded in 1972 as a manufacturer
of gear measuring machines and instru-
ments. Yuzakis continuing commitment
to the development of innovative prod-
ucts that contribute to gear quality im-
provement worldwide is evidenced by his
companys many and diverse offerings
matched to ever-changing global mea-
surement needs. Applications have been filed for international
patent protection of the NDG Method measuring machine. Yu-
zaki is a member of the Measurement Committee of the Japan
Gear Manufacturers Association (JGMA) and a member of the
Japan Society of Mechanical Engineers (JSME).

60 GEARTECHNOLOGY March/April 2011 www.geartechnology.com

You might also like